|
Telemedicine and e-Health
Lifelink: 3G-Based Mobile Telemedicine System
To cite this article:
Christian Alis, Carlos del Rosario, Bernardino Buenaobra, Carlo Mar Blanca.
Telemedicine and e-Health.
April 2009,
15(3): 241-247.
doi:10.1089/tmj.2008.0098.
Published in Volume: 15 Issue 3: April 21, 2009
Christian Alis, M.S.National Institute of Physics, University of The Philippines, Diliman, Quezon City, The Philippines. Carlos del Rosario, Jr., B.S.National Institute of Physics, University of The Philippines, Diliman, Quezon City, The Philippines. Bernardino Buenaobra, M.S.National Institute of Physics, University of The Philippines, Diliman, Quezon City, The Philippines. Carlo Mar Blanca, Ph.D.National Institute of Physics, University of The Philippines, Diliman, Quezon City, The Philippines. Lifelink, a mobile real-time telemonitoring and diagnostic facility, linked by mobile phones, was evaluated. Thirty 2-hour electrocardiogram (ECG) signals from cardiac patients were analyzed using detrended fluctuation analysis. This analysis detects long-range correlations embedded in non-stationary temporal series. The patient population was categorized in three classes: healthy, congestive heart failure, and atrial fibrillation. This system permits a physician to evaluate the mobile phone–based system as a novel approach to analyze ECG at a distance. Current wired telemedicine systems encounter difficulties when implemented in archipelagic developing countries because of the high cost of fixed infrastructure. In this research, we devised Lifelink, a mobile real-time telemonitoring and diagnostic facility to command and control remote medical devices through mobile phones. The whole process is phone-based, effectively freeing offsite medical specialists from stationary monitoring consoles and endowing the system with the potential to increase the number participating consultants. The electrocardiogram (ECG) readings are analyzed using a detrended fluctuation technique and classified into pathological cases using an unassisted K-means clustering algorithm. We analyzed 30 batches of 2-hour ECG signals taken from cardiac patients (20 males, 10 females, mean age 46.7 years) with pre-diagnosed pathologies. The method successfully categorized the 30 subjects without user intervention into the following cases: normal (at 86.7% accuracy), congestive heart failure (86.7%), and atrial fibrillation (80.0%). The synergy of mobile monitoring and fluctuation analysis presents a powerful platform to reach remote, underserved communities with poor or nonexistent wired communication structures. It is likely to be essential in the development of new mobile diagnostic and prognostic measures. 
|
|